Ok, another “I can’t connect to the internet” question, but with the difference that I really really can’t, try as I might, with ONE of my computers (running Opensuse 11.2)
In many ways it should be easy because, I have ANOTHER box which is connected to the net via a static ip address, also running 11.2.
So I disconnect the cable from the successfully connected box to the unconnected one, and then I register the exact same network settings from one to the other.
This includescgoing through the dialogs in yast and double checking in /etc/sysconfig/network. routes, default gateways, static ip, broadcast, mask, dns server ips … all this is the exact same.
When I connect the cable ifconfig gives a RUNNING status with the correct static ip … but, no successful pings to major sites … just returns immediately with an unknown host reply.
I’ve been through the process three times now, and I have to be missing something, maybe somethign obvious … and that’s why I post here …any thoughts from anybody?
The second unsuccessfully connected box usually works prefectly, the eth0 port has no problems (it usually connects to private lan).
I reckon I have covered everything: any suggestions from anybody?
Are you using NetworkManager or ifup? if you are using NetworkManager
the file /etc/resolv.conf has the correct DNS servers or has nothing?
VampirD
General Failure is the supreme commander of the Microsoft army.
All operation made by this army ends on him.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.12 (GNU/Linux)
Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org/
I think the cure, is what emileneth says … telnet all the different components. I will try this and report back.
I use ifup, not NetworkManager. In fact, what I was suspicious of, in all my troubleshooting, was that there was an extra opensuse service which was deflecting or impeding all my manual network configurations.
Ok, the days have passed by and I haven’t foudn a solution to this … no inet access with this static ip configuration.
following emileneth’s advice … I can ping to the gateway but not telnet on port 80.
I get “connection refused” for this.
Hm… I looked up “route” and it seems to be there:
xxx.xxx.xxx.0
That comes up, though my gw is xxx.xxx.xxx.1
but I guess that’s not a problem … I can ping, so that doesn’t seem to be the weak spot (in /etc/sysconfig/network/routes the correct gateway is given).
I also ran into this ifplugd daemon, this sounds to be a “suse-thing” I haven’t seen it before. In any case, it sems to only prioritise one of my nw interfaces … ad it’s prioritisinght right one, so no problems there.
In any case, the telnettest seems to point to a http problem, that the gateway doesn’t like it.
I have no firewall running … so, well, I expect, no blocking there.
Would welcome any more clues from anybody, thanks.
On 12/15/2009 01:06 PM, stabu wrote:
>
> Ok, the days have passed by and I haven’t foudn a solution to this …
> no inet access with this static ip configuration.
>
> following emileneth’s advice … I can ping to the gateway but not
> telnet on port 80.
>
> I get “connection refused” for this.
>
> Hm… I looked up “route” and it seems to be there:
> xxx.xxx.xxx.0
>
> That comes up, though my gw is xxx.xxx.xxx.1
>
> but I guess that’s not a problem … I can ping, so that doesn’t seem
> to be the weak spot (in /etc/sysconfig/network/routes the correct
> gateway is given).
>
> I also ran into this ifplugd daemon, this sounds to be a “suse-thing” I
> haven’t seen it before. In any case, it sems to only prioritise one of
> my nw interfaces … ad it’s prioritisinght right one, so no problems
> there.
>
> In any case, the telnettest seems to point to a http problem, that the
> gateway doesn’t like it.
>
> I have no firewall running … so, well, I expect, no blocking there.
Is your gateway running an httpd server? If not, you got exactly the
response expected. Does ‘telnet www.samba.org 80’ get anywhere? How
about the following:
haha, I never thought to use traceroute, cos I’ve always viewed it as a “fun” tool … i.e. something to show the different nodes your working inet connection goes through.
I’ll try your suggestions out tomorrow.
Ref.
Is your gateway running an httpd server? If not, you got exactly the response expected.
That’s interesting … no, no httpd running there at my gateway. But, my “connection refused” should be expected? Hmmm…
I followed lwfinger’s and Xerionth’s advice and got
> traceroute 216.83.154.106
traceroute to 216.83.154.106 (216.83.154.106), 30 hops max, 40 byte packets using UDP
Unable to connect to 216.83.154.106: Network is unreachable
> traceroute www.samba.org
www.samba.org: Name or service not known
> telnet www.samba.org
telnet: www.samba.org: Name or service not known
> traceroute 59.144.169.16
traceroute to 59.144.169.16 (59.144.169.16), 30 hops max, 40 byte packets using UDP
Unable to connect to 59.144.169.16: Network is unreachable
My impression from these results is that traceroute doesn’t say a whole lot at my stage of the troubleshooting. Maybe at the next stage, it could be more useful.
So what stage am I at? I get pings from my gateway, but “connection refused” when I try telnet 80 on it.
All this started when I tried connecting this network interface to a small private LAN of mine (also static ip) which was not connected to the internet.
Before I that event, I had full internet access on static ip with the connected wide area LAN.
So previously I had not problems, and I had used yast all the way, and it was able to fully sort out the internet connection.
Now, on my other network interface, eth1, I have a wireless connection which used DHCP. I have no problem with this. I execute the following script on it:
#!/bin/bash
# script to set wlan connectivity via DHCP
ifconfig eth1 down
iwconfig eth1 essid "<whateverithappenstobe>"
ifconfig eth1 up
iwconfig eth1 enc open <CODE>
dhcpcd -t 15 -d eth1
and try giving the eth0 device a static IP. Let’s say your router is 192.168.1.1, then set eth0 to have 192.168.1.71, netmask to 255.255.255.0, gateway to 192.168.1.1, and use Google’s public nameservers (8.8.8.8 and 4.4.8.8 these are IP-addresses, I use them, they’re fast) for DNS.
the “route add default gw”, that solved it! That was way off my radar!
I admit to having run the “route” command several times and seeing:
Kernel IP routing table
43 Destination Gateway Genmask Flags Metric Ref Use Iface
44 169.12.48.0 * 255.255.255.0 U 0 0 0 eth0
45 link-local * 255.255.0.0 U 0 0 0 eth0
46 loopback * 255.0.0.0 U 0 0 0 lo
where 169.12.48.1 was my gateway . my untrained eye thought this readout was OK. I also admit to thinking that, by running yast networking settings beforehand, “route”'s setting would be taken care of. It would seem to me now, that yast network settings aren’t hugely sensitive to “route” (though they did work in the beginning when I installed Opensuse). Soe clearly, when I followed Xerionth’s advice and stuck in the gw in the route command I got
Kernel IP routing table
43 Destination Gateway Genmask Flags Metric Ref Use Iface
44 169.12.48.0 * 255.255.255.0 U 0 0 0 eth0
45 link-local * 255.255.0.0 U 0 0 0 eth0
46 loopback * 255.0.0.0 U 0 0 0 lo
47 default 169.12.48.1 0.0.0.0 UG 0 0 0 eth0
And then everything came good.
Kudos to Xerionth …thanks very much. I’ll admit to changing the settings on resolv.conf too, though I did not stick in the gateway ip there … that seemed unconventional.
SO my lesson from this is … if network is out … make sure you get resolv.conf and route settings right. Make sure you know how to interpret their status! Thanks to Xerionth, but also to all others who made suggestions!
On 12/17/2009 12:46 PM, Xerionth wrote:
>
> Glad to hear it worked out
>
> And idd, normally this should be taken care of by yast but there seems
> to be a bug in this version of opensuse which messes up the routing.
It is fixed. All one needs to do is update, then delete the old
connection and remake it.
